The Stanford v Roche decision was not at all about the proper technical steps to make Bayh-Dole into a vesting statute. Even the Court’s minority opinion–what the lawyer-krakkens fixated on–was a musing on whether there should be any difference in the equitable ownership of an invention whether the ownership was promised in the future (“I agree to assign”) or assigned now for future inventions (“I hereby assign”). Both seem to accomplish the same result, and both have uncertainties. The first promises to assign once an invention has been identified and meets the conditions specified. The second assigns an utterly undefined thing that does not exist and becomes effective only when an invention is made and is found to meet the conditions specified. But the issue in Stanford v Roche was not about the form of the assignment, but the fact that an equitable interest had been conveyed and nothing in Bayh-Dole voided that conveyance.
Look at the scenario. University has a patent policy that says inventors own whenever possible. The university only requires ownership inventions when it must by contract. A university lab director is on the technical advisory board of a company. He wants his post-doc to work at the company for nine months to learn its new technology. The company agrees, on the condition that anything the post-doc invents as a result of his work at the company is assigned to the company. Post-doc agrees, and signs an agreement that says he will assign and hereby does assign all such inventions to the company. The university agrees with this arrangement.
At this point, everything follows the university’s patent policy. The university has no ownership interest in any invention that the post-doc might make as a result of his work at the company. The post-doc has agreed that the company has the equitable interest he may control in any such invention. The post-doc then returns to the university and immediately reports as an invention just what he and the company were working on when he was at the company.
The university ignores the post-doc’s agreement with the company, files patent applications and later obtains an assignment from the post-doc and from his co-inventors. When the patents issue, the university demands that the company license the patents for a product that the company is already selling, and when the company refuses, the university sues. Much fussing follows. The first court finds for the university. The appeals court finds for the company. Here’s the question that went to the Supreme Court:
Whether a federal contractor university’s statutory right under the Bayh-Dole Act, 35 U.S.C. §§ 200-212, in inventions arising from federally funded research can be terminated unilaterally by an individual inventor through a separate agreement purporting to assign the inventor’s rights to a third party.
Answer: No. But the university’s right to own isn’t statutory. And since that was the question, and that was the answer, the Supreme Court didn’t get around to teaching a Reading 101 for Dummies course in what Bayh-Dole does require, how it operates, and where to look for the requirements that a university takes on in accepting federal money to support faculty research proposals.
The Supreme Court held that Bayh-Dole did not give a federal contractor university any statutory right to take ownership of inventions. It merely gave federal contractor universities the right to retain ownership of an invention, once it had ownership, relative to a federal agency’s request for control of the invention. Thus, no individual inventor could possibly “terminate unilaterally” the university’s rights in any such invention by assigning it to someone else, since the university had no rights under Bayh-Dole in the invention until the university properly obtained assignment of the invention. The court advised Stanford that if it wanted to obtain ownership of such inventions, a well-established method was to have patent agreements with employees obligating them to assign their inventions to the university.
Lawyer-krakkens pounced on this advice as if a change in magical wording would have changed the outcome of the case. Hardly. It would have made no difference whether Stanford had asked its personnel to “hereby assign” all inventions that Stanford was required to own. The Supreme Court made it clear that not only did Bayh-Dole not vest ownership of federally supported inventions with Stanford, but that Stanford was not required by law to own those inventions. So a present assignment would not have done anything under Stanford’s patent policy because the conditions that would make that assignment effective did not exist. It is not the manner of the obligation to assign, but its scope, that matters. The invention in dispute was outside the scope of Stanford’s policy claim. Thus, a present assignment would not have operated at the time the invention was reported to Stanford.
The obligation to assign to Roche would have operated, however. That was a valid contract to convey the post-doc’s entire interest in his invention to Roche. That, in fact, is what the Appeals Court had found. When Stanford got around to obtaining the post-doc’s assignment, the post-doc had nothing to assign, having already obligated his rights in the invention to Roche. Even if Stanford had changed its policy and required its own present assignment to future inventions the moment the post-doc returned to Stanford, it would not have made any difference in the outcome. In the case of competing assignments, 35 USC 261 controls:
An interest that constitutes an assignment, grant or conveyance shall be void as against any subsequent purchaser or mortgagee for a valuable consideration, without notice, unless it is recorded in the Patent and Trademark Office within three months from its date or prior to the date of such subsequent purchase or mortgage.
To make void a prior assignment, Stanford could not have “notice” of the prior assignment and must have provided “valuable consideration” for the assignment. But Stanford did have notice of the prior assignment–so the court found. And at any rate, Stanford argued that the inventor had to assign–by law–and so the assignment obtained by Stanford did not involve paying “a valuable consideration” to the inventor.
As for priority–who assigned first–Stanford would have had to require the post-doc to assign future inventions in the area of the company’s work and arising from access to company information before the post-doc agreed to the company’s terms. But as soon as Stanford approved (accepted, did not contest, was indifferent to) the company’s conditions for the post-doc to work for months at the company and gain access to the company’s technology, how could Stanford also claim that its prior requirement still was effective? Stanford would have approved an exception to its own requirement to assign when it approved the post-doc’s deal with the company. To do otherwise would be equivalent to intending to steal the company’s future inventions. This is just the bad behavior that companies guard against by requiring visitors to agree to assign their inventions made that arise from their access to the company’s technology and resources.
But Stanford had no claim to priority with its own statement of assignment obligations because the invention that was made by the post-doc was out of scope for the university’s requirement to assign. So there was no first promise to assign that was later trumped by a present assignment. There was no obligation under Stanford’s patent policy to assign the invention in question, ever. The first valid obligation to assign was the one that the post-doc made to the company, with Stanford’s blessing. That the invention as reported was attributed to a federally funded project at Stanford does not change the post-doc’s valid obligation to assign to the company. That’s what the Supreme Court ruled. Bayh-Dole does not affect the ownership of inventions made with federal support until a party to the federal funding agreement owns the invention. Then the standard patent rights clause requirements kick in. Thus, there was no way for the post-doc to “void” Stanford’s rights in inventions made with federal support. Nor could Bayh-Dole be construed to require Stanford to own inventions made with federal support.
But university patent administrators took the Stanford v Roche case to claim that to comply with federal law, they had to require not merely a promise to assign inventions but a present assignment of future inventions–none of which has anything to do with the question posed by Stanford v Roche. A university complies with the standard patent rights clause by requiring the (f)(2) written agreement, designating personnel responsible for patent matters, educating employees with regard to the importance of timely disclosure of inventions, and reporting inventions to the government when these inventions are reported to the university.